一种基于迭代抑制谱泄漏的改进相位差算法

由于非同步采样所带来的频谱泄漏效应是影响相位差算法电参量测量精度的主要原因,利用高阶窗的旁瓣特性可抑制频谱泄漏,但二次谐波幅值的估计精度仍难以显著提升,并且由于高阶窗所带来频率分辨率损失的原因,通常需要增加待测信号的采样窗长,不利于频率波动和实时性要求较高的场合实现高精度测量。为此,提出一种加sine窗函数的改进型相位差算法,利用峰值谱线间相位差估计出的相对频偏,计算出除被测谐波分量之外的其他分量的长程谱泄漏对峰值谱线的干扰之和并将其减去后,重新利用其峰值谱线间的相位差计算相对频偏,循环迭代估计。基于该方法,与现有加窗相位差算法分别在无噪声、有噪声以及频率波动环境下进行仿真比较。实验结果表明,该方法迭代矫正计算4次即可,能有效抑制频谱泄露效应,减小估计偏差,提高电网电参量测量精度和实时性,能够满足电网测量需要。

An improved phase difference algorithm based on iteration of suppressing spectrum leakage

The spectrum leakage effect caused by non-synchronous sampling is the main impact for the measurement precision of the phase difference algorithm, The spectral leakage can be suppressed by the side lobe characteristic of high-order window function, However, the estimation accuracy of the second harmonic amplitude is still difficult to improve significantly, And the reason for the loss of frequency resolution due to high Window function, The sampling window is usually increased, It is not conducive to the frequency fluctuation and the real time requirementSis higher. In this paper, a modified phase difference algorithm with sine window is presented, The relative frequency deviation is estimated by the phase difference between peak spectral lines, and calculate the sum of the long range spectrum leakage of the components other than the measured harmonic components,And subtract it from the measured component, we update relative frequency deviation by the phase difference between peak spectral lines, cyclic iterative estimation. Based on this method, Compared with the existing phase difference algorithm in noise-free and noise environment and frequency fluctuation environment. Experimental results show that this method can be used for the iterative correction of 4 times, can effectively suppress the spectrum leakage effect, reduced estimation error,and improve the measurement precision of power grid electrical parameters, this method can be used for grid measurement requirements.